Dynamic Object Tracking

1. A method of tracking an object in an environment using a moving camera, comprising: acquiring an earlier image and a later image of the environment from an image stream captured by the camera; identifying later environment features located in the environment in the later image, earlier environment features located in the environment in the earlier image, and earlier object features located on the object in the earlier image; determining a camera movement from the earlier image to the later image using a difference in location between the earlier environment features and the later environment features; estimating object features in the later image using the earlier object features and the determined camera movement; locating, in the later image, matched object features that are actual object features in the later image at a same location as the estimated object features; determining that the object has moved between the earlier image and the later image if a number of matched object features does not exceed a threshold; and determining that the object has not moved between the earlier image and the later image if the number of matched object features exceeds the threshold.

2. The method of claim 1 , further comprising: displaying a virtual image based on motion of the object if the object has moved.

3. The method of claim 2 , further comprising, if the object has moved: determining a pose of the object in the later image using the actual object features in the later image and the earlier object features; and updating a location of the virtual image based on the determined pose of the object.

4. The method of claim 1 , further comprising, if the object leaves a field of view of the camera: relocating the object in an image in the image stream if the object re-enters the field of view.

5. The method of claim 1 , further comprising: determining a pose of the object before the earlier image is acquired, by locating object features stored in a memory in an image in the image stream, wherein in the identifying earlier object features and earlier environment features, the pose of the object is used to distinguish between the earlier object features and the earlier environment features.

6. The method of claim 1 , wherein the earlier environment features and earlier object features are identified by distinguishing between environment features and object features in an image in the image stream, further comprising: generating a collection of object features by obtaining additional object features after distinguishing between environment features and object features.

7. The method of claim 1 , wherein the later image is a current image in the image stream, and the earlier image is acquired from a memory.

8. The method of claim 1 , wherein the camera and display are components of one of: a head mounted display, a smartphone, and a tablet.

9. A head mounted display that tracks an object in an environment, comprising: a camera that captures an image stream of the environment, including an earlier image and a later image; a processor configured to: identify later environment features located in the environment in the later image, earlier environment features located in the environment in the earlier image, and earlier object features located on the object in the earlier image; determine a camera movement from the earlier image to the later image using a difference in location between the earlier environment features and the later environment features; estimate object features in the later image using the earlier object features and the determined camera movement; locate, in the later image, matched object features that are actual object features in the later image at a same location as the estimated object features; determine that the object has moved between the earlier image and the later image if a number of matched object features does not exceed a threshold; and determine that the object has not moved between the earlier image and the later image if the number of matched object features exceeds the threshold.

10. The head mounted display of claim 9 , further comprising: a display that displays a virtual image based on motion of the object if the object has moved.

11. The head mounted display of claim 10 , wherein the processor is further configured to, if the object has moved: determine a pose of the object in the later image using the actual object features in the later image and the earlier object features; and update a location of the virtual image based on the determined pose of the object.

12. The head mounted display of claim 9 , wherein the processor is further configured to, if the object leaves a field of view of the camera: relocate the object in an image in the image stream if the object re-enters the field of view.

13. The head mounted display of claim 9 , wherein the processor is further configured to: determine a pose of the object before the earlier image is acquired, by locating object features stored in a memory in an image in the image stream, wherein in the identifying earlier object features and earlier environment features, the pose of the object is used to distinguish between the earlier object features and the earlier environment features.

14. The head mounted display of claim 9 , wherein the processor: identifies the earlier environment features and earlier object features by distinguishing between environment features and object features in an image in the image stream, and is further configured to generate a collection of object features by obtaining additional object features after distinguishing between environment features and object features.

15. The head mounted display of claim 9 , further comprising: a memory, wherein the later image is a current image in the image stream, and the earlier image is acquired from the memory.

16. A non-transitory computer readable medium that embodies instructions that cause one or more processors to perform a method comprising: acquiring an earlier image and a later image of an environment from an image stream captured by a camera; identifying later environment features located in the environment in the later image, earlier environment features located in the environment in the earlier image, and earlier object features located on the object in the earlier image; determining a camera movement from the earlier image to the later image using a difference in location between the earlier environment features and the later environment features; estimating object features in the later image using the earlier object features and the determined camera movement; locating, in the later image, matched object features that are actual object features in the later image at a same location as the estimated object features; determining that the object has moved between the earlier image and the later image if a number of matched object features does not exceed a threshold; and determining that the object has not moved between the earlier image and the later image if the number of matched object features exceeds the threshold.

17. The non-transitory computer readable medium of claim 16 , wherein the method further comprises: instructing a display to display a virtual image based on motion of the object if the object has moved.

18. The non-transitory computer readable medium of claim 17 , wherein the method further comprises, if the object has moved: determining a pose of the object in the later image using the actual object features in the later image and the earlier object features; and updating a location of the virtual image based on the determined pose of the object.

19. The non-transitory computer readable medium of claim 16 , wherein the method further comprises, if the object leaves a field of view of the camera: relocating the object in an image in the image stream if the object re-enters the field of view.

20. The non-transitory computer readable medium of claim 16 , wherein the method further comprises: determining a pose of the object before the earlier image is acquired, by locating object features stored in a memory in an image in the image stream, wherein in the identifying earlier object features and earlier environment features, the pose of the object is used to distinguish between the earlier object features and the earlier environment features.